Alveolar macrophage cytotoxicity for normal lung fibroblasts ismediated by nitric oxide release

Daniel L. Morgan*, Cassandra J. Shines

Respiratory Toxicology, Laboratory of Molecular Toxicology, National Institute of Environmental Health Sciences, PO Box 12233,

Research Triangle Park, NC, USA

Received 9 July 2003; received in revised form 13 August 2003; accepted 19 August 2003

Abstract

Nitric oxide (NO) released by activated alveolar macrophages (AM) can mediate effects on target cells and can also react withsuperoxide anion (O2-) to form peroxynitrite (PN), a highly cytotoxic product. The role of NO and PN in AM cytotoxicity for

normal lung cells was investigated using co-cultures of rat lung fibroblasts (FB) and rat AM treated with lipopolysacchar-ide+interferon-g (LI). AM and FB, alone and in co-culture, were treated with LI for 5 days and cell viability measured. The culturemedia was analyzed for NO, TNF-a, O2-, and IL-1b. A decreased FB viability was correlated with increased NO release by LI-activated AM. Pretreatment of co-cultures with the inducible NOS inhibitor l-NAME caused dose-related decreases in NO release

by AM and increases in FB viability. Although TNF-a release was increased in co-cultures treated with LI, the viability of FB wasnot affected when cultures were treated with similar concentrations of TNF-a in the absence of AM. O2- could not be detected inthe media and addition of superoxide dismutase (SOD) did not protect FB. These data suggest that neither O2- nor PN contributed

to the loss of cell viability. Activated AM may kill normal rat lung FB through a NO-mediated pathway that does not involve PN.Published by Elsevier Ltd.

Keywords: Alveolar macrophage; Lung fibroblast; Lipopolysaccharide; Cytotoxicity; Nitric oxide

1. Introduction

Interactions between activated alveolar macrophages(AM) and other lung cells are implicated in the pro-gression of many inflammatory pulmonary diseases(Sibille and Reynolds, 1990). Alveolar macrophages arean essential line of defense for the airways and alveolarsurfaces; however, they can also cause injury to the lung(Brain, 1986). Inhalation of particulates and other for-

eign material can activate AM resulting in the increasedproduction and release of a number of potent inflam-matory mediators. Most mediators released by activatedAM are directed at specific target cell receptors andcause a controlled change in target cell function. How-ever, mediators such as reactive oxygen and nitrogenspecies are nonspecific in their action and can inad-vertently cause injury to surrounding lung cells. Inflam-mation and progressive damage to normal tissues ischaracteristic of chronic obstructive pulmonary diseasesand may be due in part to reactive oxygen and nitrogenspecies released by activated AM.Lipopolysaccharide (LPS), a proinflammatory com-

ponent of Gram-negative bacterial membranes, is aubiquitous contaminant present in airborne particulatematter such as cotton dusts (Christiani et al., 1993),organic dusts (Smid et al., 1992) and in metal workingfluid aerosols (Gordon, 1992). Chronic exposure to LPSis associated with the development and/or progressionof many pulmonary diseases including asthma, chronicbronchitis, and progressive irreversible airflow obstruc-tion, that are all characterized by chronic inflammatory

0887-2333/$ - see front matter Published by Elsevier Ltd.

doi:10.1016/j.tiv.2003.08.007

Toxicology in Vitro 18 (2004) 139–146

www.elsevier.com/locate/toxinvit

Abbreviations: AM, rat alveolar macrophage; EMEM, Earle’s

minimal essential medium; FB, rat lung fibroblast; FCS, fetal calf

serum; HBSS, Hank’s balanced salts solution; IFN-g, interferon-g;iNOS, inducible nitric oxide synthase; IL-1b, interleukin-1b; LPS,

lipopolysaccharide; LI, lipopolysaccharide+interferon-g; l-NAME,

N-nitro-l-arginine-methyl ester; MTT, 3-[4,5-dimethylthiazol-2-yl]-

2,5-diphenyltetrazolium bromide; NaNO2, sodium nitrite; NO, nitric

oxide; O2�, superoxide anion; SOD, superoxide dismutase; PN, perox-

ynitrite; PBS, phosphate buffered saline; TNF-a, tumor necrosis fac-tor-a.* Corresponding author. Tel.: +919-541-2264; fax+1-919-541-

0356.

E-mail address: morgand@niehs.nih.gov (D.L. Morgan).

processes in the lung (Vernooy, et al., 2002). LPS sti-mulates inflammatory responses by activating macro-phages and other cells of the immune system. Previousstudies have shown that LPS-activated AM releaseincreased amounts of tumor necrosis factor (TNF-a)(Decker et al., 1987; Ichinose et al., 1988; Brandolini etal., 2001), and interleukin-1 (IL-1b) (Decker et al., 1987;Brandolini et al., 2001). These two cytokines are con-sidered to be the primary mediators of inflammation inthe lung (Ulrich et al., 1991).Activation of rat AM with LPS also causes an

increased production and release of nitric oxide (NO)(Pace et al., 1994; Gao et al., 1998), as well as super-oxide anion (O2

�) (Watson et al., 1994). Although themechanism(s) by which NO causes lung injury is notclear, the reaction of NO with O2

� to form the highlyreactive peroxynitrite (PN) may be involved (Radi et al.,1991). Peroxynitrite has potent oxidizing propertiestoward biological macromolecules including proteinand nonprotein sulfhydryls, DNA, and membranephospholipids (Radi et al., 1991; Kooy et al., 1995;King et al., 1992). In addition, PN specifically reactswith free or cell-associated protein tyrosines (Ischiro-poulos et al., 1992). Nitration of cellular tyrosine resi-dues by PN can inhibit tyrosine phosphorylation andthereby interfere with important tyrosine kinase medi-ated signaling pathways (Mandoro et al., 1997; Gow etal., 1996; Kong et al., 1996). Tyrosine nitration by PNhas been detected in human acute lung injury (Kooy etal., 1995), and in the lungs of laboratory animals treatedwith LPS (Wizemann et al., 1994).Alveolar macrophage and lung cell interactions are

likely the result of complex networks between differentcell types in the lung and many different interactingcytokines and reactive oxygen and nitrogen species.Because of the difficulties in unraveling these complex invivo interactions, an in vitro model consisting of AMand one target cell type (lung fibroblast) was developed.This co-culture system was used to evaluate the poten-tial nonspecific effects of NO released by activated AMon a normal lung target cell.

2. Materials and methods

2.1. Reagents

Tumor necrosis factor-a (TNF-a), Elisa Kit #10019;interleukin-1b (IL-1b) (colorimetric Elisa Kit #RLB00,and interferon-g (IFN-g) were purchased from R&DSystems (Minneapolis, MN). Lipopolysaccharide (LPS;PN L3755), 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenylte-trazolium bromide (MTT; PN M2128); sodium nitrite(NaNO2; PN S2252); Griess reagent (PN G4410); N-nitro-l-arginine-methyl ester (l-NAME; PN N5751);fetal calf serum (FCS; F2442), and superoxide

dismutase (SOD; PN S2515) were purchased fromSigma Chemical Co. (St. Louis, MO). Polyclonal anti-body to 3-nitrotyrosine was obtained from UpstateBiotechnology (Lake, Placid, NY), and the secondaryantibody, goat anti-mouse IgG, Avidin-Biotin-horseradish peroxidase complex, and DAB-Ni substrate werepurchased from Vector Laboratories (Burlingame, CA).

2.2. Rat lung fibroblasts (FB)

Rat lung FB were selected as the target cell in this co-culture model because they are from the same speciesand tissue as rat AM, and are normal (non-trans-formed) cells that can be easily isolated from the lungand passaged in high purity. Rat lung FB were isolatedfrom lungs of adult Sprague–Dawley rats by enzymaticdigestion and cultured in Earle’s minimal essentialmedium (EMEM) containing 5% FCS as described byKumar et al. (1988). FB were grown to confluence andpassaged at least 3 times before use in these experi-ments.

2.3. Rat alveolar macrophages (AM)

Rat AM were collected by bronchoalveolar lavage(BAL) of male, Sprague–Dawley rats (250 g bodyweight; Charles River, Raleigh, NC). Rats were eutha-nized by pentobarbital overdose and thoracotomy. Thethoracic cavity was exposed and the pulmonary vascu-lature was perfused via the pulmonary artery with coldsterile saline. The trachea was surgically exposed andcannulated with a blunt 16-gauge needle and the lungswere gently inflated with about 10-ml sterile, cold iso-tonic saline. The saline was gently withdrawn from thelung and the BAL fluid placed in a 15-ml centrifuge tubeon ice. This process was performed 3-times for each rat.The BAL fluid was centrifuged (600�g) to pellet cellsand the cells from all animals were combined andresuspended in EMEM+5% FCS. Cell counts weremade using a hemocytometer.

2.4. FB and AM co-culture

Lung FB were plated at 2�104/well in 24-well dishesin EMEM+5% FCS and incubated overnight. FB cul-tures were rinsed 3� with 37 �C HBSS to remove unat-tached cells then freshly lavaged AM (105) were platedin each well containing FB. AM were allowed to attachfor 2 h, then unattached cells were rinsed off and freshEMEM+5% FCS with or without (controls) 15 mg/mllipopolysaccharide+50 U/ml interferon-g (LI) wasadded to each co-culture (Day 1). In a preliminary dose-response experiment, concentrations of LPS greater than15 mg/ml decreased the viability of FB to an unacceptablelevel, and LPS concentrations less than 15 mg/ml pro-duced sub-optimal activation of AM (as measured by

140 D.L. Morgan, C.J. Shines / Toxicology in Vitro 18 (2004) 139–146

target cell toxicity). The addition of interferon resultedin an increased activation of AM without increasing thetoxicity of LPS for FB in the absence of AM. Thecombination of LPS and interferon-g has been shown tomore rapidly induce NO formation than LPS alone(Gao et al., 1998). Co-cultures were incubated at 37 �Cand on day 5 the supernatants were collected and cellviability measured. In preliminary experiments, theeffects of LI-activated AM on FB viability were eval-uated after 1, 3, 5, and 7 days. The effects of AM on FBviability increased with time and attained a maximumeffect at day 5 (data not shown). All experiments wereconducted in duplicate using at least 4 cultures/treatment.

2.5. MTT viability assay

Cell viability was determined by the MTT assay asmodified by Kasugai et al. (1990). The MTT assayassesses mitochondrial function and is generally con-sidered to be an estimation of cell viability. Co-cultureswere rinsed 1� with HBSS to remove detached cells(>98% FB). Detached FB were unable to take up and/or metabolize the dye and were considered nonviable.After removing detached cells, 500 ml EMEM+5%FCS containing 50 ml of 5 mg/ml MTT was added toeach well. After incubating for 4 h at 37 �C, the mediawas removed and 500 ml dimethylsulfoxide was added toeach well. The cultures were incubated at room tem-perature for 10 min, then 100 mL of solution was trans-ferred from each well to a corresponding well on a 96-well plate. The absorbance was read on a plate reader at562 nm (Bio-Tek Instruments, Winooski, VT).AM alone and in co-culture did not take up the MTT

under any treatment conditions, and did not contributeto the color formation. For this reason any color for-mation in AM-FB co-cultures was attributed to MTTuptake by viable FB. In order to verify cell viability,AM cultures and co-cultures were stained with TrypanBlue and observed microscopically.

2.6. Nitric oxide assay

Nitric oxide production was estimated by measuringthe accumulation of the stable NO metabolite, nitrite, inthe culture supernatants using the Griess assay (Greenet al., 1982). The supernatants from cultures were cen-trifuged at about 600�g for 10 min at room temper-ature to remove debris and detached cells. A standardcurve was prepared using NaNO2 in water with a rangeof 0–100 mm. Aliquots (50 ml) of standards or sampleswere transferred to 96-well plates and mixed with 25 mlof 1% sulfanilamide in 5% phosphoric acid followed by25 ml of 0.1% naphthylethylenediamine hydrochloride.After incubating at room temperature for 15 min, theabsorbance was read at 562 nm on a plate reader (Bio-Tek Instruments, Winooski, VT). All samples were

analyzed in duplicate. The limit of quantification was2.5 mm nitrite.

2.7. l-NAME

On day 1 co-cultures were rinsed 1� with HBSS andthen fresh medium containing LI and either 0, 0.625,1.25, 2.5 or 5.0 mm l-NAME. Co-cultures were incu-bated for 5 days at 37 �C, then the supernatants werecollected from each well and analyzed for NO asdescribed above. The viability of the attached cells inthe co-cultures was determined using the MTT assay.

2.8. Superoxide anion assay

Superoxide anion levels in medium from LI-activatedAM were determined on days 1 and 5 by the methods ofPick and Mizel (1981). Briefly, the cultures were treatedwith media with and without (controls) LI. After LItreatment for either 1 or 5 days, the media was replacedwith EMEM containing horse heart ferricytochrome C(75 mm). After incubation for about 5 min, the culturemedia was transferred to 96-well dishes and the oxid-ation of ferricytochrome C was measured on a platereader at 562 nm.

2.9. Superoxide dismutase (SOD) protection assay

On day 1 co-cultures were rinsed 1� with HBSS andfresh medium containing LI with or without SOD (75,100, 150, or 250 mg/ml) was added to the wells. Cultureswere incubated for 5 days at 37 �C. Because the SODactivity may decrease with time in the culture media,fresh SOD was added to the culture media daily. Cul-tures were then rinsed 1� with HBSS and cell viabilitywas measured using the MTT assay.

2.10. Media transfer

Although most AM were attached to FB in co-cul-ture, it was not clear whether direct cell-to-cell contactwas necessary for toxicity, or if the toxicity was medi-ated by factors released by AM into the culture media.AM cultures were treated with LI for 5 days, then theculture media was removed and filtered (0.2 micron) toremove cells and debris, and then added to FB cultures.Control FB cultures were treated with either fresh cul-ture medium (EMEM+5% FCS), or medium from AMcultured for 5 days in the absence of LI. After treatmentfor 5 days, the medium was removed and cell viabilityevaluated using the MTT assay.

2.11. TNF-� assay

Supernatants from the AM, FB and co-cultures werecollected on day 5 and were then assayed for TNF-� by

D.L. Morgan, C.J. Shines / Toxicology in Vitro 18 (2004) 139–146 141

ELISA (R&D Systems) following the manufacturersinstructions. The minimal detectable level of this assaywas 5.1 pg TNF-a/ml.

2.12. IL-1�

Supernatants from the LI-treated and control AM,FB and co-cultures were collected on day 5 and werethen assayed for IL-1b by ELISA (R&D Systems) fol-lowing the manufacturers instructions. The minimaldetectable level of this assay was 5.0 pg IL-1b/ml.

2.13. Immunostaining for nitrotyrosine

AM and FB co-cultures were plated on glass coverslips in 6-well dishes in order to facilitate staining ofcells and microscopic observation. Co-cultures wereplated and treated for 5 days with medium with andwithout (controls) LI as described above. On day 5, cells

were rinsed with PBS, then fixed with fresh 4% glutar-aldehyde in PBS+0.1% Tween 20 for 5 min and airdried. The fixed cells were incubated with polyclonalanti-nitrotyrosine antibody (1:500) in PBS for 1 hour,then rinsed with PBS and incubated 30 min in biotiny-lated secondary antibody solution. Cells were thenincubated 30 min with the avidin-biotin-horse radishperoxidase complex followed by a 30-minute incubationwith DAB-Ni substrate. The cover slips were mountedon glass slides and the cells were observed by lightmicroscopy.

3. Results

3.1. Viability

In untreated AM-FB co-cultures, the FB grew rapidlyand were near confluence by day 5. Macrophagesattached to FB with no apparent adverse effects on FB(Fig. 1A). In co-cultures treated with LI, the FB becamespindle-shaped and by day 5 many had detached fromthe culture dish (Fig. 1B). The viability of the attachedFB was measured following co-culture with AM, treat-ment with LI, and following co-culture with AM andtreatment with LI (Fig. 2). Co-culture with AM did notsignificantly affect the viability of FB relative to controls(FB treated with fresh media only). Treatment of FBcultures with LI resulted in about a 20% decrease inviability. Treatment of AM-FB co-cultures with LIresulted in a much greater decrease in FB viability (77%nonviable).

3.2. Nitric oxide release

Only low concentrations of NO (<5 mM) were pre-sent in the culture media from untreated FB or AMwhen cultured alone or in co-culture (Fig. 3). Treatmentof FB cultures with LI had no significant effect on NOrelease. However, treatment of AM with LI resulted in asignificant (P<0.05) 7-fold increase in NO release.Similarly, treatment of co-cultures with LI resulted in asignificant (P<0.05) 5-fold increase in NO release.

Fig. 2. Effects of LI-activated alveolar macrophages (AM) on lung

fibroblast (FB) viability. Values represent means�sem (n=4). * Sig-

nificantly less than viability of untreated FB cultures (P<0.05).

Fig. 1. (A) Normal appearance of untreated rat alveolar macrophages

(arrows) and rat lung fibroblasts in co-culture (day 5) (10�). (B)

Alveolar macrophage (arrows) and rat lung fibroblast co-culture on

day 5 after treatment with LI (lipopolysaccharide/interferon). (10�).

142 D.L. Morgan, C.J. Shines / Toxicology in Vitro 18 (2004) 139–146

There was no significant difference in NO release fromLI-treated AM and LI-treated co-cultures.

3.3. NO synthase inhibition

In order to determine the role of NO in FB toxicity,co-cultures were treated with the inducible NO synthaseinhibitor l-NAME. l-NAME caused a concentration-related decrease in the release of NO into the co-culturemedium (Fig. 4). Treatment with 5 mm l-NAME resul-ted in about a 90% reduction in NO release. Cell viabi-lity was measured in the co-cultures following l-NAMEtreatment (Fig. 5). Cell viability in the co-culturesincreased with increasing concentrations of l-NAME.

3.4. Superoxide anion

The culture media from LI treated co-cultures wasanalyzed for superoxide anion. If present, the con-centrations of O2

� were below the limit of quantification(data not shown). As a further check, superoxide dis-mutase (SOD) was added to the co-culture media toreduce any O2- present in the media. Only slight, non-significant (P>0.05) changes were observed in viabilityof cultures treated with SOD (data not shown).

3.5. Media transfer

In order to determine whether direct cell-to-cell con-tact was necessary for toxicity, FB were treated witheither control medium (fresh EMEM+5% FCS), cul-ture medium from untreated AM, or culture mediumfrom LI-treated AM. FB viability was significantly(P<0.05) reduced only in cultures treated with mediafrom LI-treated AM (Fig. 6).

3.6. Tumor necrosis factor-�

The concentration of TNF-a was measured in mediafrom FB and AM cultured alone or in co-culture, andwith or without LI treatment (Fig. 7). In the absence ofLI, relatively low concentrations of TNF-a (30 to 45 pg/ml) were detected in the media of FB and AM culturedalone or co-cultured. Treatment of FB with LI had noeffect on concentrations TNF-a in culture medium(P>0.05). However, LI treatment of AM alone or in co-culture resulted in statistically significant increases(P<0.05) in TNF-a release. A maximum of about 65pg/ml TNF-a was measured in the media from AMcultures treated with LI. To determine the extent towhich TNF-a contributed to the decreased cell viability

Fig. 4. Inhibition of nitric oxide (NO) release by treatment of co-

cultures with the iNOS inhibitor l-NAME prior to LI treatment.

Values represent means�SEM (n=4). *Significantly less than control

(no l-NAME); (P<0.05).

Fig. 3. LI-stimulated release of nitric oxide (NO) by alveolar macro-

phages (AM). Values represent means�SEM (n=4). *yGroups with

different superscripts are significantly different (P<0.05).

Fig. 5. Inhibition of LI-activated alveolar macrophage (AM) effects

on fibroblast (FB) viability by pretreatment with the iNOS inhibitor

l-NAME. Values represent means�SEM (n=4). *Significantly

greater than control (no l-NAME); (P<0.05).

Fig. 6. Effects of alveolar macrophage (AM)-conditioned medium

on fibroblast (FB) viability. Values represent means�SEM (n=8).

* Significantly less than viability of FB cultures treated with control

media (P<0.05).

D.L. Morgan, C.J. Shines / Toxicology in Vitro 18 (2004) 139–146 143

in LI-treated co-cultures, AM and FB co-cultures weretreated with 65 pg/ml of TNF-a in the absence of LI.TNF-a had no significant (P>0.05) effect on cell viabilityin treated co-cultures (data not shown).

3.7. Interleukin-1-� (IL-1�)

The concentrations of IL-1b were measured in mediafrom FB and AM cultured alone or in co-culture, andwith or without LI treatment. Treatment with LI had nosignificant effect (P>0.05) on the levels of IL-1breleased into the media (data not shown). Concentra-tions of IL-1b in the culture media ranged from 208 to240 pg/ml with and without LI-activation.

3.8. Immunostaining for nitrotyrosine

Co-cultures were immunostained for nitrotyrosineafter treatment for 5 days with media with and without(controls) LI. The stained cells were observed micro-scopically for the presence of nitrotyrosine. There wasno evidence of nitrotyrosine formation in either controlsor LI-treated cells (data not shown).

4. Discussion

Chronic inflammation and progressive damage tonormal tissues is characteristic of chronic obstructivepulmonary diseases and may be due in part to reactiveoxygen and nitrogen species released by activated AM.Many mediators released by AM are directed at specifictarget cell receptors and cause a controlled change incell function. However, mediators such as reactive oxy-gen and nitrogen species are nonspecific in their actionand cause injury to surrounding normal lung cells. Inthis study we investigated the role of NO in the cyto-toxicity of LI-activated AM for normal lung FB.Lung fibroblasts play an important role in a number

of chronic obstructive pulmonary diseases. Rat lung FB

were used as the target cells in this model because anormal (nontransformed) cell type that was easily iso-lated from the lung and passaged in high purity wasdesired. It was also necessary to use primary cells fromthe rat because transformed and tumor target cells, andtarget cells from different species would be recognized as‘‘foreign’’ by rat AM and could result in cell inter-actions that do not occur with normal target cells.Although lung FB are typically not in direct contactwith AM, they are still a relevant target cell since theyare in contact with interstitial lung macrophages (Brodyet al., 1992). More importantly, this study demonstratedthat the toxicity was mediated by a soluble factorreleased into the medium, and direct contact betweenthe AM and FB target cell was not necessary.Nitric oxide is considered to be important in pul-

monary inflammation and tissue repair, although itsprecise role is not clear (Belvisi et al., 1995). Previousstudies have suggested that macrophage-generated NOis involved in LPS-induced tissue injury (Wright et al.,1992; Nathan, 1992) and in most of these studies toxi-city was attributed to highly reactive PN produced bythe reaction of NO with O2

�. While the results of thepresent study also demonstrated that NO released byactivated AM was involved in the cytotoxicity for ratlung FB, the mechanism did not appear to involve O2

�

or PN. LI-activated AM released significantly increasedlevels of NO; however, O2

� was not detected, and theaddition of superoxide dismutase (SOD) to the co-cul-ture medium did not protect the FB. Because both NOand O2

� are required for PN formation (Radi et al.,1991) these results indicated that cytotoxicity may notinvolve PN. In support of these data, immunostainingfor N-tyrosine on FB was negative (data not shown)indicating that nitration of membrane tyrosines by PNdid not occur. Alveolar macrophage toxicity for lung FBclearly required NO, since inhibition of NO synthesisinhibited the toxicity for FB.Similar results were reported by Hirano (1996) who

demonstrated that SV40-transformed rat type II pul-monary epithelial (SV40T2) cells were lysed after co-culture with LPS-activated rat AM. As in the presentstudy, the SV40T2 target cells were protected by aniNOS inhibitor, but not by SOD, indicating that toxicityinvolved NO released by the activated AM and not PN.Similarly, Fricker et al. (1999) demonstrated that LPS-activated murine macrophages (RAW 264 cell line) werecytotoxic towards P815 murine mastocytoma cells, andthat addition of a NO synthase inhibitor exerted a con-centration-dependent protective effect. Addition ofSOD had no effect on macrophage NO production orcell killing, leading the authors to conclude that tumorcell killing was mediated primarily by NO and did notinvolve macrophage derived O2

�.Nitric oxide is involved in inflammation and there is

growing evidence that NO released by activated AM

Fig. 7. TNF-a release in alveolar macrophage (AM) and fibroblast

(FB) cultures treated with LI. Values represent means�SEM (n=4).

*y{Groups with different superscripts are significantly different

(P<0.05).

144 D.L. Morgan, C.J. Shines / Toxicology in Vitro 18 (2004) 139–146

plays an important role in pulmonary diseases, althoughthe mechanism is unclear. While the results of the cur-rent study implicate NO in the cytotoxic mechanism,neither O2-, nor PN appear to be involved in the toxi-city. The cytotoxicity of NO has been suggested toresult from inhibition of target cell respiration andDNA synthesis (Lancaster and Hibbs, 1990; Hibbs,1993). The proposed mechanism for these effectsinvolves nitrosylation of nonheme iron of enzymes in themitochondrial electron transport chain, and ribonucleo-tide reductase. Because NO is not a nitrosylating agent,these mechanisms would require formation of otherreactive nitrogen species that are capable of nitrosylation.Other data indicate that NO or other reactive nitrogenintermediates may target cytochrome oxidase therebyinhibiting oxidative metabolism (Cleeter et al., 1994).Nitric oxide synthase (NOS) activity has been shown

to be significantly increased in lungs from patientswith inflammatory lung diseases such as asthma,bronchiolitis obliterans, and cystic fibrosis (Belvisi etal., 1995). Inhibition of inducible NOS provided sig-nificant protection of FB against AM cytotoxicity;however, protection was not complete suggesting thatother cytotoxic mediators may be involved. TNF-a andIL-1b are potent inflammatory mediators released byactivated macrophages that could contribute to FBcytotoxicity. In this study only low levels of IL-1b werereleased by AM with and without LI activation. How-ever, TNF-a release by AMwas significantly increased byLI activation. Levels up to 65 pg/ml were present in theco-culture media after LI treatment. Yet in the absence ofAM, treatment of FB with up to 100 pg/ml TNF-a hadno effect on FB viability. It is possible that NO may actsynergistically with other mediators released by LPS-activated AM. Several studies have indicated that NOand TNF-a produce a synergistic toxicity or that NOmayserve as an intermediary and signal the release of second-ary mediators by AM (Lorsbach et al., 1993; Higuchi etal., 1990). Activated AM release a number of potentiallycytolytic mediators, in addition to TNF-a and IL-1b. Thepotential role of other AM mediators in FB cytotoxicityis an area for additional research.In summary, rat lung FB viability was significantly

decreased in the presence of LI-activated AM. Activa-tion of AM resulted in a significant increase in therelease of nitric oxide (NO) and TNF-a; however, onlyNO was correlated with FB toxicity. Pretreatment of co-cultures with a NO synthase inhibitor protected FB.The absence of superoxide anion in the medium fromactivated AM, and the inability of exogenous super-oxide dismutase to protect FB from activated AM indi-cate that toxicity may not be due to peroxynitriteformed from NO and superoxide anion. Activation ofAM resulted in toxicity for normal FB by a mechanisminvolving NO but a role for peroxynitrite could not bedemonstrated.

Acknowledgements

The authors thank Drs. James Bonner and Dori Ger-molec for critical review of this manuscript and H. Priceand D. Crawford for technical assistance.

References

Belvisi, M., Barnes, P.J., Larkin, S., Yacoub, M., Tadjkarimi, S.,

Williams, T.J., 1995. Nitric oxide synthase activity is elevated in

inflammatory lung disease in humans. European Journal of Phar-

macology 283, 255–258.

Brain, J.D., 1986. Toxicological aspects of alterations of pulmonary

macrophage function. Annual Review of Toxicology 26, 547–565.

Brandolini, L., Intilangelo, A., Caselli, G., Bertini, R., 2001. Role of

tumor necrosis factor-alpha in endotoxin-induced lung parenchymal

hyporesponsiveness in mice. Biochemical Pharmacology 62, 1141–

1144.

Brody, A.R., Bonner, J.C., Overby, L.H., Badgett, A., Kalter, V.,

Kumar, R.K., Bennett, R.A., 1992. Interstitial pulmonary macro-

phages produce platelet-derived growth factor that stimulates rat

lung fibroblast proliferation in vitro. Journal of Leukocyte Biology

51, 640–648.

Christiani, D.C., Velazques, A., Wilcox, M., Olenchock, S.A., 1993.

Airborne endotoxin concentrations in various work areas within a

cotton mill in Central America. Environmental Research 60, 187–

192.

Cleeter, M.W.J., Cooper, J.M., Darley-Usmar, V.M., Moncada, S.,

Schapira, A.H.V., 1994. Reversible inhibition of cytochrome C

oxidase, the terminal enzyme of the mitochondrial respiratory

chain, by nitric oxide. Federation of European Biochemical Society

Letters 345, 50–54.

Decker, T., Lohmann-Matthes, M.L., Gifford, G., 1987. Cell asso-

ciated tumor necrosis factor (TNF) as a killing mechanism of acti-

vated cytotoxic macrophages. Journal of Immunology 138, 957–962.

Fricker, S.P., Slade, E., Powell, N.A., 1999. Macrophage-derived

superoxide is not required for nitric oxide mediated tumour cell

killing by RAW 264 cells. Anticancer Research 19, 553–556.

Gao, J.J., Filla, M.B., Fultz, M.J., Vogel, S.N., Russell, S.W., Mur-

phy, W.J., 1998. Autocrine/paracrine IFN-ab mediates the lipopo-

lysaccharide-induced activation of transcription factor Stat1a in

mouse macrophages: Pivotal role of Stat1a in induction of the

inducible nitric oxide synthase gene. Journal of Immunology 161,

4803–4810.

Gordon, T., 1992. Acute respiratory effects of endotoxin-con-

taminated machining fluid aerosols in guinea pigs. Fundamental

and Applied Toxicology 19, 117–123.

Gow, A.J., Duran, D., Malcolm, S., Ischiropoulos, H., 1996. Effects of

peroxynitrite-induced protein modification of tyrosine phosphor-

ylation and degradation. Federation of European Biochemical

Society Letters 385, 63–66.

Green, L.C., Wagner, D.A., Glogowski, J.S., Skipper, P.L., Wishnok,

J.S., Tannenbaum, S.R., 1982. Analysis of nitrate, nitrite, and [15-

N]nitrite in biological fluids. Analytical Biochemistry 126, 131–138.

Hibbs Jr, J.B., 1993. Cytokine induced synthesis of nitric oxide from

l-arginine: a cytotoxic mechanism that targets intracellular iron. In:

Reiderer, P., Youdim, M.B.H. (Eds.), Iron in central nervous sys-

tem disorders. Key topics in brain research. Vienna-Springer Verlag,

pp. 155–171.

Higuchi, M., Higashi, N., Taki, H., Osawa, T., 1990. Cytolytic

mechanisms of activated macrophages. Tumour necrosis factor and

l-arginine-dependent mechanisms act synergistically as the major

cytolytic mechanisms of activated macrophages. Journal of Immu-

nology 144, 1425–1431.

D.L. Morgan, C.J. Shines / Toxicology in Vitro 18 (2004) 139–146 145

Hirano, S., 1996. Interaction of rat alveolar macrophages with pul-

monary epithelial cells following exposure to lipopolysaccharide.

Archives of Toxicology 70, 230–236.

Ichinose, Y., Bakouche, O., Tsao, J.Y., Fidler, I.J., 1988. Tumor

necrosis factor and IL-1 associated with plasma membranes of acti-

vated human monocytes lyse monokine-sensitive but not monokine-

resistant tumor cells, whereas viable activated monocytes lyse both.

Journal of Immunology 141, 512–518.

Ischiropoulos, H., Zhu, L., Chen, J., Tsai, M., Martin, J.C., Smith,

C.D., Beckman, J.S., 1992. Peroxynitrite mediated tyrosine nitration

catalyzed by superoxide dismutase. Archives of Biochemistry and

Biophysics 298, 431–437.

Kasugai, S., Hasegawa, N., Ogura, H., 1990. A simple in vitro cyto-

toxicity test using the MTT (3-(4,5)-dimethyl thiazol-2-YL)-2,5-

diphenyl tetrazolium bromide) colorimetric assay - analysis of

eugenol toxicity on dental pulp cells (RPC-C2A). Japanese Journal

of Pharmacology 52, 95–100.

King, P.A., Anderson, V.E., Edwards, J.O., Gustafson, G., Plumb,

R.C., Suggs, J.W., 1992. A stable solid that generates hydroxyl

radical upon dissolution in aqueous solutions: reaction with pro-

teins and nucleic acid. Journal of the American Chemical Society

114, 5430–5432.

Kong, S.K., Yim, M.B., Stadtman, E.R., Chock, P.B., 1996. Perox-

ynitrite disables the tyrosine phosphorylation regulatory mechan-

ism: lymphocyte specific tyrosine kinase fails to phosphorylate

nitrated cdc2(6-20) NH2 peptide. Proceedings of the National

Academy of Sciences, USA 93, 3377–3382.

Kooy, N.W., Royall, J.A., Ye, Y.Z., Kelly, D.R., Beckman, J.S., 1995.

Evidence for in vivo peroxynitrite production in human acute lung

injury. American Journal of Respiratory and Critical Care Medicine

151, 1250–1254.

Kumar, R.K., Bennett, R.A., Brody, A.R., 1988. A homologue of

platelet-derived growth factor produced by rat alveolar macro-

phages. Federation of American Societies for Experimental Biology

Journal 2, 2272–2277.

Lancaster Jr, J.R., Hibbs Jr, J.B., 1990. EPR demonstration of iron-

nitrosyl complex formation by cytotoxic activated macrophages. Pro-

ceedings of the National Academy of Sciences, USA 87, 1223–1227.

Lorsbach, R.B., Murphy, W.J., Lowenstein, C.J., Snyder, S.H., Rus-

sell, S.W., 1993. Expression of the nitric oxide synthase gene in

mouse macrophages activated for tumor cell killing: molecular basis

for the synergy between interferon-g and lipopolysaccharide. Jour-

nal of Biological Chemistry 268, 1908–1913.

Mandoro, T.H., Shaffer, B.C., Vostal, J.G., 1997. Peroxynitrite-

induced tyrosine nitration and phosphorylation in human platelets.

Free Radicals in Biology and Medicine 22, 1005–1063.

Nathan, C., 1992. Nitric oxide as a secretory product of mammalian

cells. Federation of American Societies for Experimental Biology

Journal 6, 3051–3064.

Pace, J.L., Lowenstein, C.J., Phillips, T.A., Chen, L.-C., Morrison,

D.C., Hunt, J.S., Russell, S.W., 1994. Population dynamics of

inducible nitric oxide synthase production by LPS- and LPS/IFN-g-stimulated mouse macrophages. Journal of Endotoxin Research 1,

227.

Pick, E., Mizel, D., 1981. Rapid microassays for the measurement of

superoxide and hydrogen peroxide production by macrophages in

culture using an automatic enzyme immunoassay reader. Journal of

Immunological Methods 46, 211–226.

Radi, R., Beckman, J.S., Bush, K.M., Freeman, B.A., 1991. Perox-

ynitrite oxidation of sulfhydryls. Journal of Biological Chemistry

266, 4244–4250.

Sibille, Y., Reynolds, H.Y., 1990. Macrophages and polymorpho-

nuclear neutrophils in lung defense and injury. American Review of

Respiratory Disease 141, 471–501.

Smid, T., Heederik, D., Houba, R., Quanjer, P.H., 1992. Dust- and

endotoxin-related respiratory effects in animal feed industry. Amer-

ican Review of Respiratory Disease 146, 1474–1479.

Ulrich, T.R., Watson, L.R., Yin, S., Guo, K., Wang, P., Thang,

H., del Castillo, J., 1991. The intratracheal administration of

endotoxin and cytokines. American Journal of Pathology 138,

1485–1496.

Vernooy, J.H.J., Dentener, M.A., van Suylen, R.J., Buurman, W.A.,

Wouters, E.F.M., 2002. Long-term intratracheal lipopolysaccharide

exposure in mice results in a chronic lung inflammation and persis-

tent pathology. American Journal of Respiratory Cell and Mole-

cular Biology 26, 152–159.

Watson, R.W.G., Redmond, H.P., Bouchier-Hayes, D., 1994. Role of

endotoxin in mononuclear phagocyte-mediated inflammatory

responses. Journal of Leukocyte Biology 56, 95–103.

Wizemann, T.M., Gardner, C.R., Laskin, J.D., Quinones, S., Dur-

ham, S.K., Goller, N.L., Ohnishi, S.T., Laskin, D.L., 1994. Pro-

duction of nitric oxide and peroxynitrite in the lung during acute

endotoxemia. Journal of Leukocyte Biology 56, 759–768.

Wright, C.E., Rees, D.D., Moncada, S., 1992. Protective and patho-

logical roles of nitric oxide in endotoxin shock. Cardiovascular

Research 26, 48–57.

146 D.L. Morgan, C.J. Shines / Toxicology in Vitro 18 (2004) 139–146